n8meter 1.0.1

Fixes the day-one report from the community: local and LAN n8n instances now work out of the box.

The URL guard was tuned for hosted multi-tenant deployments and refused private addresses — but self-hosted n8meter usually runs right next to its n8n, so localhost:5678, 192.168.… and host.docker.internal:5678 are the normal case. They are allowed by default now; file:// and friends, redirect tricks and oversized responses are still refused.

• Set BLOCK_PRIVATE_URLS=1 if your n8meter is reachable by people who are not its operator — that restores the strict guard.
• compose.yaml now maps host.docker.internal to the host gateway, so the container reaches an n8n on the host machine on Linux too.

Update: git pull && npm install (or rebuild the container).

n8meter 1.0.0

First public release.

What it does: reads token usage from your n8n executions, prices it against a daily catalog of 2,500+ models, and shows spend per workflow, client and day — on your own machine.

• Per-workflow metering with per-node attribution
• Provider-billed counts where available; n8n estimates flagged with ≈
• Clients & projects, multi-scope budgets with webhook alerts
• Webhook ingest fallback for instances without API access
• SQLite, no build step, no telemetry

Quick start: git clone → npm install → npm start → open http://localhost:4780 and paste your n8n API key. Or docker compose up -d. Just looking? npm run demo.

n8meter is an independent community project, not affiliated with n8n GmbH.